Abstract A CapsNet-based Chinese character font representation model is proposed to represent Chinese character font by the representation of components. Firstly, representative vectors of all categories are generated by the model. Then, a group of component representative vectors are selected by the Euclidean-distance-based outlier detection according to component probabilities. Finally, these vectors are utilized to form the Chinese character font representations. The experimental results show that the proposed model, merely trained on component fonts, is capable of identifying components of Chinese characters and automatically generating effective representation of Chinese characters.
Fund:Supported by National Natural Science Foundation of China(No.61672158), Natural Science Foundation of Fujian Province(No.2018J1798), Industry-University Cooperation Foundation of Fujian Province(No.22018H6010)
About author:: (XIE Haiwen, master student. His research interests include image processing.)(YE Dongyi(Corresponding author), Ph.D., professor. His research interests include computational intelligence and data mining.)(CHEN Zhaojiong, master, professor. Her research interests include intelligent image processing and computational intelligence.)
[1] 刘钊. 汉字印刷字体发展、设计与应用研究.博士学位论文.北京:中央美术学院, 2007. (LIU Z.Chinese Characters Printed Font: Development, Designation and Application Research. Ph.D. Dissertation. Beijing, China: Central Academy of Fine Arts, 2007.) [2] 刘文予,万菲,朱光喜.基于形态学的新的汉字字形自动生成方法.计算机学报, 1999, 22(3): 235-240. (LIU W Y, WAN F, ZHU G X.A Novel Method of Chinese Font Composition Based on Morphology. Chinese Journal of Computers, 1999, 22(3): 235-240.) [3] 潘志庚,马小虎,张明敏,等.基于Fourier级数描述器的多种汉字字形自动生成.软件学报, 1996, 7(6): 331-338. (PAN Z G, MA X H, ZHANG M M, et al. The Fourier Descriptor Based Automatic Generation Method for Multiple Chinese Fonts. Journal of Software, 1996, 7(6): 331-338.) [4] 杨建,张明敏,石教英.基于Fourier描述器的汉字字形生成及合成的改进算法.计算机辅助设计与图形学学报, 2001, 13(7): 617-621. (YANG J, ZHANG M M, SHI J Y.An Improved Algorithm for Chinese Font Generation and Morphing Based on Fourier Descriptor. Journal of Computer-Aider Design and Computer Graphics, 2001, 13(7): 617-621.) [5] 严伟荣,蔡士杰.基于笔划特征的宋体字形衍生方法.中文信息学报, 1995, 9(1): 16-24. (YAN W R, CAI S J.An Approach to Derive SONGTI Typeface Based on Stroke Character. Journal of Chinese Information Proce-ssing, 1995, 9(1): 16-24.) [6] 潘志庚,马小虎,石教英.动态汉字库自动生成算法.自动化学报, 1996, 22(5): 561-567. (PAN Z G, MA X H, SHI J Y.The Automatic Generation Algorithm for Dynamic Chinese Font. Acta Automatica Sinica, 1996, 22(5): 561-567.) [7] 刘成东,连宙辉,唐英敏,等.基于部件拼接的高质量中文字库自动生成系统.北京大学学报(自然科学版), 2018, 54(1): 35-41. (LIU C D, LIAN Z H, TANG Y M, et al. Automatical System to Generate High-Quality Chinese Font Libraries Based on Component Assembling. Acta Scientiarum Naturalium Universitatis Pekinensis, 2018, 54(1): 35-41.) [8] MIYAZAKI T, TSUCHIYA T, SUGAYA Y, et al. Automatic Ge-neration of Typographic Font from a Small Font Subset[C/OL].[2017-12-02]. https://arxiv.org/pdf/1701.05703.pdf. [9] GATYS L A, ECKER A S, BETHGE M.Image Style Transfer Using Convolutional Neural Networks // Proc of the IEEE Confe-rence on Computer Vision and Pattern Recognition. Washington, USA: IEEE, 2016: 2414-2423. [10] JIANG Y, LIAN Z H, TANG Y M, et al. DCFont: An End-to-End Deep Chinese Font Generation System // Proc of the SIGGRAPH Asia 2017. New York, USA: ACM, 2017. DOI: 10.1145/3145749.3149440. [11] 陈宗明. 汉字符号学:一种特殊的文字编码.南京:江苏教育出版社, 2001. (CHEN Z M.Chinese Character Semiotics: A Special Type of Character Coding. Nanjing, China: Jiangsu Education Press, 2001.) [12] SABOUR S, FROSST N, HINTON G E.Dynamic Routing between Capsules[C/OL]. [2018-04-02].https: //arxiv. org/pdf/1710. 09829.pdf. [13] LECUN Y, BOTTOU L, BENGIO Y, et al. Gradient-Based Learning Applied to Document Recognition. Proceedings of the IEEE, 1998, 86(11): 2278-2324. [14] 韩布新. 部件组合-潜在的汉字结构层次.中文信息学报, 1995, 9(3): 27-32. (HAN B X.Combination of Chinese Character Constituents-A Latent Structural Unit. Journal of Chinese Information Processing, 1995, 9(3): 27-32.) [15] ABADI M, BARHAM P, CHEN J M, et al. TensorFlow: A System for Large-Scale Machine Learning // Proc of the 12th USENIX Conference on Operating Systems Design and Implementation. Berlin, Germany: Springer, 2016: 265-283. [16] KINGMA D P, BA J L. ADAM: A Method for Stochastic Optimization[C/OL]. [2018-04-02]. https://arxiv.org/pdf/1412.6980.pdf.
2019, Vol. 32 
